This collaboration has developed a bipolar disorder family resource that has proven to be uniquely valuable in testing clinically-based hypotheses about the genetic heterogeneity of the illness. We have ascertained and assessed about 150 nuclear families with a bipolar I proband and two or more siblings with a major affective disorder. The analysis of clinical features in the ascertained families has revealed important patterns, including bipolar II, psychotic bipolar disorder, and panic disorder comorbidity. One important goal is to derive and test genetically meaningful phenotypic subtypes. Tests of familial aggregation of variables and factors are also being performed, and positive results will be followed by covariate analyses of genetic linkage and association data. A new genome-wide linkage scan has been performed by the Center for Inherited Disease Research (CIDR) on 105 families not previously scanned. An additional 112 families will be scanned in the 5th and final year of this project. The data will be analyzed with both standard linkage methods and with conditional logistic methods that allow for use of covariates such as clinical variables or factors, or other loci. We have also performed association studies through SNP genotyping (using the Illumina platform) in 4 chromosomal regions: 6q21-23, 13q31-33, 18q21-22 and 22q12. Analyses of the SNP data have revealed several signals that will be followed up with replication in an independent sample. We previously reported results of an association study of close to 5000 SNP markers on chromosomes 13, 18, and 22, as well as in selected candidate genes. Preliminary results implicated several genes in bipolar disorder that are good candidates for replication in independent samples. We have found evidence that risk of bipolar disorder in these families is associated with genetic variation in the gene encoding FKBP5, genes encoding proteins involved in the Wnt signaling pathway, and genes that influence circadian rhythms. We have also generated a database of clinical variables, The Bipolar Disorder Phenome Database, based in part on this sample. The database contains several hundred clinical variables, harmonized with those collected by the NIMH Genetics Initiative, describing the course, symptoms, and clinical picture of bipolar disorder. We have made the database available to the scientific community to support research into the genetics of bipolar disorder and related conditions. For example, we used these data to demonstrate that postpartum mood symptoms aggregate in bipolar disorder pedigrees, suggesting that these symptoms may help define a familial subtype of the disorder. Over the past year, in collaboration with scientists at Cold Spring Harbor Laboratory, we have used cutting-edge genetic tools to search for relatively small structural variations in chromosomes, known as copy number variants(CNVs), that may play a role in bipolar disorder. Using these tools, we found evidence that some CNVs strongly over-represented among people with schizophrenia may also play a role in bipolar disorder. Ongoing work is focussed on detecting de novo CNVs that may give rise to bipolar disorder, especially in an early-onset form, among the offspring of parents who do not suffer from the condition.